Search Results

Search found 8224 results on 329 pages for 'sometimes'.

Page 284/329 | < Previous Page | 280 281 282 283 284 285 286 287 288 289 290 291  | Next Page >

• Read header data from files on remote server

  - by rejeep

  Hi! I'm working on a project right now where I need to read header data from files on remote servers. I'm talking about many and large files so I cant read whole files, but just the header data I need. The only solution I have is to mount the remote server with fuse and then read the header from the files as if they where on my local computer. I've tried it and it works. But it has some drawbacks. Specially with FTP: Really slow (FTP is compared to SSH with curlftpfs). From same server, with SSH 90 files was read in 18 seconds. And with FTP 10 files in 39 seconds. Not dependable. Sometimes the mountpoint will not be unmounted. If the server is active and a passive mounting is done. That mountpoint and the parent folder gets locked in about 3 minutes. Does timeout, even when there's data transfer going (guess this is the FTP-protocol and not curlftpfs). Fuse is a solution, but I don't like it very much because I don't feel that I can trust it. So my question is basically if there's any other solutions to the problem. Language is preferably Ruby, but any other will work if Ruby does not support the solution. Thanks!

  Read the article

• Is "tip-of-the-day" good?

  - by Jonta

  Many programs (often large ones, like MS Office, The GIMP, Maxthon) have a feature called "tip-of-the-day". It explains a small part of the program, like this one in Maxthon: "You can hide/show the main menu bar by pressing Ctrl+F11" You can usually browse through them by clicking next. And other options provided are "Previous", "Close", "Do not show at startup". I think I like the way Maxthon used to handle this; in the browser's statusbar (down at the bottom usually, together with "Done", the progress-bar etc), there would sometimes be a small hint or tip on what else you could do with it. As Joel Spolsky wrote in his article-series "User Interface Design for Programmers", people don't like reading manuals. But we still want them to use the program, and the features they could benefit from, don't we? Therefore, I think it is useful to have such a feature, without the annoyance of the pop-up on startup. What do you think? Pop-up? Maxthonstyle? No way?

  Read the article

• How to gracefully handle YouTube thumbnail broken images when videos are taken down?

  - by mrjf

  Our site contains some lists of YouTube videos with their thumbnails linked to the movies themselves. We fetch the thumbnail URLs from the YouTube API, and hotlink to them in situ on YouTube's servers. So our image tags look like: <img src="http://i.ytimg.com/vi/o6CHqSN7O-o/2.jpg" alt="" width="133" height="78" /> The problem is that sometimes a video gets removed -- and so does the thumbnail. We don't know when that might happen, and our thumbnails just turn into broken images. How can we handle this? There are a number of solutions: Download thumbnails and store them locally -- movie won't work, but that's ok, it will explain it was removed, and we avoid the broken image Check periodically with the API to see if the image thumbnail has changed -- if it no longer exists, substitute our own "movie removed" thumbnail. This is pretty heavy on the API calling! Use JavaScript to replace broken images (don't like this one much) Our ideal solution would be to point the img src to a location on YouTube that would display a friendly "move removed" image when a movie goes down. However, that doesn't seem to exist. Anyone else dealt with this? Thanks!

  Read the article

• Looking for out-of-place directories in an SVN working copy?

  - by jthg

  An annoyance that I sometimes come across with SVN is the working copy getting corrupted by one of the .svn folders getting moved from its original location. It doesn't happen often if you're careful and use the proper tools for all moves and renames, but it still somehow happens from time to time. First, does anyone know if there's a good way to catch the problem before a commit is even done? Cruise control usually catches the problem, but there are plenty of cases it wouldn't catch. Second, is there a quick and easy way to check for out-of-place .svn folder if I suspect that there is one? I can definitely do it manually by deducing what directory is out of place based on the compiler errors or by diffing the working copy with another clean checkout. But, this seems like a problem that SVN can diagnose in a second by giving me a list of all directories whose parent directory in the working copy doesn't match its parent directory in the repository. There there some way to have SVN give me a list like that? Thanks.

  Read the article

• How can I simplify this user interface?

  - by Bears will eat you

  I'm writing an internal-tools webapp; one of the central pages in this tool has a whole bunch of related commands the user can execute by clicking one of a number of buttons on the page, like this: Ideally, all of the buttons would fit on one line. Ordinarily I'd do this by changing each widget from a button with a (sometimes long) text label to a simple, compact icon - e.g. could be replaced by a familiar disk icon: Unfortunately, I don't think I can do this for every button on this particular page. Some of the command buttons just don't have good visual analogs - "VDS List". Or, if I needed to add another button in the future for some other kind of list, I'd need two icons that both communicate "list-ness" and which list. So, I'm still considering this option, but I don't love it. So it's come time for me to add yet another button to this section (don't you love internal tools?). There's not enough room on that single line to fit the new button. Aside from the icon solution I already mentioned, what would be a good* way to simplify/declutter/reduce or otherwise improve this UI? *As per Jakob Nielsen's article, I'd like to think that a dropdown menu is not the solution.

  Read the article

• PHP SDK Requires two logins...doesn't recognize first

  - by Jay Konieczny

  I am using the following code to authenticate whether users are logged in or not. While users can log in, they have to click the login button twice. Additionally, sometimes even after they click the log-in button twice, my "user info" part of the page (earlier in the page than the content) shows them as logged out while the actual page shows them as logged in. Here is the code. Could someone suggest a better way of handling log-ins? function isLoggedIn($facebook) { if (isset($facebook) and $facebook->getUser() != 0) { // UserID exists, but user may still not be logged in. Let's check: try { $facebook->api('/me', 'GET'); // If this succeeds, then they are logged in. return true; } catch(FacebookApiException $e) { // Some kind of error, so not logged in. if(session_id() === '') session_destroy(); return false; } } else { if(session_id() === '') session_destroy(); return false; } } Thanks!

  Read the article

• GTK+: How do I process RadioMenuItem choice without marking it chosen? And vise versa

  - by eugene.shatsky

  In my program, I've got a menu with a group of RadioMenuItem entries. Choosing one of them should trigger a function which can either succeed or fail. If it fails, this RadioMenuItem shouldn't be marked chosen (the previous one should persist). Besides, sometimes I want to set marked item without running the choice processing function. Here is my current code: # Update seat menu list def update_seat_menu(self, seats, selected_seat=None): seat_menu = self.builder.get_object('seat_menu') # Delete seat menu items for menu_item in seat_menu: # TODO: is it a good way? does remove() delete obsolete menu_item from memory? if menu_item.__class__.__name__ == 'RadioMenuItem': seat_menu.remove(menu_item) # Fill menu with new items group = [] for seat in seats: menu_item = Gtk.RadioMenuItem.new_with_label(group, str(seat[0])) group = menu_item.get_group() seat_menu.append(menu_item) if str(seat[0]) == selected_seat: menu_item.activate() menu_item.connect("activate", self.choose_seat, str(seat[0])) menu_item.show() # Process item choice def choose_seat(self, entry, seat_name): # Looks like this is called when item is deselected, too; must check if active if entry.get_active(): # This can either succeed or fail self.logind.AttachDevice(seat_name, '/sys'+self.device_syspath, True) Chosen RadioMenuItem gets marked irrespective of the choose_seat() execution result; and the only way to set marked item without triggering choose_seat() is to re-run update_seat_menu() with selected_seat argument, which is an overkill. I tried to connect choose_seat() with 'button-release-event' instead of 'activate' and call entry.activate() in choose_seat() if AttachDevice() succeeds, but this resulted in whole X desktop lockup until AttachDevice() timed out, and chosen item still got marked.

  Read the article

• What access token should i use?

  - by user548458

  I made a posting scores. It worked normaly sometimes. But, now I cant post any score. If I use user accessToken: string path = meID + "/scores"; var parameters = new Dictionary<string, object> (){ { "score", score.ToString () }, { "access_token", accessToken } }; facebook.post (path, parameters, ( error, obj ) => I get error: {"error":{"message":"(#15) This method must be called with an app access_token.", "type":"OAuthException", "code":15}} If I use an app access token: string path = meID + "/scores"; var parameters = new Dictionary<string, object> () { { "score", score.ToString () }, { "access_token", appAccessToken } }; facebook.post (path, parameters, ( error, obj ) => I get other error: {"error":{"message":"A user access token is required to request this resource.", "type":"OAuthException", "code":102}} Help me please, what am i doing wrong? PS: I worked well recently, but now - dont. Cant explain it...

  Read the article

• Zend Framework - How do make a hierarchy without it being a module?

  - by Josh

  Here is my specific issue. I want to make an api level which then under that you can select which method you will use. For example: test.com/api/rest test.com/api/xmlprc Currently I have api mapping to a module directory. I then setup a route to make it a rest route. test.com/api is a rest route, but I would rather have it be test.com/api/rest. This would allow me later to add others. In Bootstrap.php: $front = Zend_Controller_Front::getInstance(); $router = $front->getRouter(); $route = new Zend_Controller_Router_Route( 'api/:module/:controller/:id/*', array('module' =>'default') ); $router-addRoute('api', $route); $restRoute = new Zend_Rest_Route($front, array(), array( 'rest' )); $router-addRoute('rest', $restRoute); return $router; In application.ini: resources.frontController.moduleDirectory = APPLICATION_PATH "/modules" I know it will involve routes, but sometimes I find the Zend Framework documentation to be a little hard to follow. When I go to test.com/rest/controller/ it works how it should, but if I go to test.com/api/rest/ it tells me that my module is api and controller is rest.

  Read the article

• A very interesting MYSQL problem (related to indexing, million records, algorithm.)

  - by terence410

  This problem is pretty hard to describe and therefore difficult to search the answer. I hope some expert could share you opinions on that. I have a table with around 1 million of records. The table structure is similar to something like this: items{ uid (primary key, bigint, 15) updated (indexed, int, 11) enabled (indexed, tinyint, 1) } The scenario is like this. I have to select all of the records everyday and do some processing. It takes around 3 second to handle each item. I have written a PHP script to fetch 200 items each time using the following. select * from items where updated unix_timestamp(now()) - 86400 and enabled = 1 limit 200; I will then update the "updated" field of the selected items to make sure that it wont' be selected again within one day. The selected query is something like that. update items set updated = unix_timestamp(now()) where uid in (1,2,3,4,...); Then, the PHP will continue to run and process the data which doesn't require any MYSQL connection anymore. Since I have million records and each record take 3 seconds to process, it's definitely impossible to do it sequentially. Therefore, I will execute the PHP in every 10 seconds. However, as time goes by and the table growth, the select getting much slower. Sometimes, it take more than 100 seconds to run! Do you guys have any suggestion how may I solve this problem?

  Read the article

• Common protected data member in base class?

  - by EXP0

  I have a base class and several derived classes. The derived classes use some common data, can I just put those common data as protected member of the base class? I know the protected member breaks encapsulation sometimes, so I wonder if there is any good approach. Here is a specific example: class Base{ public: virtual void foo() = 0; void printData(); protected: std::vector<std::string> mData; } class Dr1 : public Base{ public: virtual void foo(); //could change mData } class Dr2 : public Base{ public: virtual void foo(); //could change mData } If I put mData into Dr1 and Dr2 as private member, then I need to put it in both of them, and I can not have printData() in Base since printData() need access to mData unless I make printData() virtual and have identical function in both Dr1 and Dr2, which doesn't make much sense to me. Is there a better way to approach this without using protected member? Thank you.

  Read the article

• Another thread safe queue implementation

  - by jensph

  I have a class, Queue, that I tried to make thread safe. It has these three member variables: std::queue<T> m_queue; pthread_mutex_t m_mutex; pthread_cond_t m_condition; and a push and pop implemented as: template<class T> void Queue<T>::push(T value) { pthread_mutex_lock( &m_mutex ); m_queue.push(value); if( !m_queue.empty() ) { pthread_cond_signal( &m_condition ); } pthread_mutex_unlock( &m_mutex ); } template<class T> bool Queue<T>::pop(T& value, bool block) { bool rtn = false; pthread_mutex_lock( &m_mutex ); if( block ) { while( m_queue.empty() ) { pthread_cond_wait( &m_condition, &m_mutex ); } } if( !m_queue.empty() ) { value = m_queue.front(); m_queue.pop(); rtn = true; } pthread_mutex_unlock( &m_mutex ); return rtn; } Unfortunately there are occasional issues that may be the fault of this code. That is, there are two threads and sometimes thread 1 never comes out of push() and at other times thread 2 never comes out of pop() (the block parameter is true) though the queue isn't empty. I understand there are other implementations available, but I'd like to try to fix this code, if needed. Anyone see any issues? The constructor has the appropriate initializations: Queue() { pthread_mutex_init( &mMutex, NULL ); pthread_cond_init( &mCondition, NULL ); } and the destructor, the corresponding 'destroy' calls.

  Read the article

• How to properly manage a complex DB structure?

  - by errr

  Let's say you have several systems using the same DB - each uses several schemes (sometimes same as the other). This structure of these schemes is somewhat very big and complicated. Now, how could you possibly manage such scheme structure? Obviously using some sort of "configuration" - the simplest would be SQL scripts, but a more reasonable solution would be XMLs which can be easily converted into SQL, or some other readable solution (for example, JPA's XMLs or Annotations). This solution though, causes a problem where you can't really tell if your configuration matches the structure of the DB schemes exactly. You can't say if those two are synchronized. Why wouldn't they? Well, in such big structure there are going to be many changes, and you won't always remember to save/commit your configuration after you've altered the schemes, or maybe you did save/commit it, but eventually didn't altered anything in the schemes and forgot to undo the changes to the configuration. More than that, another problem (not caused by the configuration, but isn't addressed by it either) is versioning. I don't see any good way of managing the DB schemes versions (say our last alteration makes 3 systems crash - not good, how to "rollback"?). And thoughts? thx.

  Read the article

• Implementing comparision operators via 'tuple' and 'tie', a good idea?

  - by Xeo

  (Note: tuple and tie can be taken from Boost or C++11.) When writing small structs with only two elements, I sometimes tend to choose a std::pair, as all important stuff is already done for that datatype, like operator< for strict-weak-ordering. The downsides though are the pretty much useless variable names. Even if I myself created that typedef, I won't remember 2 days later what first and what second exactly was, especially if they are both of the same type. This gets even worse for more than two members, as nesting pairs pretty much sucks. The other option for that is a tuple, either from Boost or C++11, but that doesn't really look any nicer and clearer. So I go back to writing the structs myself, including any needed comparision operators. Since especially the operator< can be quite cumbersome, I thought of circumventing this whole mess by just relying on the operations defined for tuple: Example of operator<, e.g. for strict-weak-ordering: bool operator<(MyStruct const& lhs, MyStruct const& rhs){ return std::tie(lhs.one_member, lhs.another, lhs.yet_more) < std::tie(rhs.one_member, rhs.another, rhs.yet_more); } (tie makes a tuple of T& references from the passed arguments.) Edit: The suggestion from @DeadMG to privately inherit from tuple isn't a bad one, but it got quite some drawbacks: If the operators are free-standing (possibly friends), I need to inherit publicly With casting, my functions / operators (operator= specifically) can be easily bypassed With the tie solution, I can leave out certain members if they don't matter for the ordering Are there any drawbacks in this implementation that I need to consider?

  Read the article

• understanding and implementing Boids

  - by alphablender

  So I'm working on porting Boids to Brightscript, based on the pseudocode here: http://www.kfish.org/boids/pseudocode.html I'm trying to understand the data structures involved, for example is Velocity a single value, or is it a 3D value, ie velocity={x,y,z} It seems as if the pseudocode seems to mix this up where sometimes it has an equation that incudes both vectors and single-value items: v1 = rule1(b) v2 = rule2(b) v3 = rule3(b) b.velocity = b.velocity + v1 + v2 + v3 If Velocity is a tripartite value then this would make sense, but I'm not sure. So my first question here is, is this the correct datastructure for a single boid based on the Pseudocode on that page: boid={position:{px:0,py:0,pz:0},velocity:{x:0,y:0,z:0},vector:{x:0,y:0,z:0},pc:{x:0,y:0,z:0},pv:{x:0,y:0,z:0}) where pc=perceived center, pv= perceived velocity I"ve implemented a vector_add, vector_sub, vector_div, and vector boolean functions. The reason I'm starting from this pseudocode is I've not been able to find anything else that is as readable, but it still leaves me with lots of questions as the data structures are not explicitly defined for each variable. (edit) here's a good example of what i'm talking about: IF |b.position - bJ.position| < 100 THEN if b.position - b[j].position are both 3D coordinates, how can they be considered "less than 100" unless they are < {100,100,100} ? Maybe that is what I need to do here, use a vector comparison function?

  Read the article

• Reversing strings in a vector using for_each and bind

  - by fmuecke

  Hi! I was wandering how it's possible to reverese strings that are contained in a vector using a single for_each command just in one "simple" line. Yea, I know it is easy with a custom functor, but I can't accept, that it can't be done using bind (at least I couldn't do it). #include <vector> #include <string> #include <algorithm> std::vector<std::string> v; v.push_back("abc"); v.push_back("12345"); std::for_each(v.begin(), v.end(), /*call std::reverse for each element*/); Edit: Thanks a lot for those funtastic solutions. However, the solution for me was not to use the tr1::bind that comes with the Visual Studio 2008 feature pack/SP1. I don't know why it does not work like expected but that's the way it is (even MS admits that it's buggy). Maybe some hotfixes will help. With boost::bind everything works like desired and is so easy (but sometimes relly messy:)). I really should have tried boost::bind in the first place...

  Read the article

• Why does my binding break down on SilverLight ProgressBars?

  - by Bill Jeeves

  I asked a similar question about charts but I have given up on that and I am using progress bars instead. Essentially, I have ten progress bars in a Silverlight control. Each is showing a different value and updating every couple of seconds (it's a process monitor). Each progress bar has the same minimum value and maximum value so the bars can be compared. Trying to follow the M-V-VM model, I have bound the value of each bar to a property in my ViewModel. All of the maximum values for the bar are bound to a single property. When the model updates, the values and the maximum can all update. This allows the bars to re-scale as the sizes grow. I'm finding that the binding will stop working sometimes on one or more bars. I suspect it is because a bar's value becomes higher than the maximum occasionally. This is because if I update the maximums first and they are going down, the values will be to high. If I update the values first when the maximum needs increasing, the values are too high again. Is there a way to stop this behaviour? Some way, perhaps, to tell the progress bars that it's OK to temporarily go too high? Or some way to tell the bindings that they shouldn't be disabled when this happens? Or maybe I've got this completely wrong and there's some other issue with ProgressBar binding I don't know about?

  Read the article

• VS2010 + IE8 Debugging woes - Element not found

  - by Chin

  I am having great difficulty trying to debug with vs2010 and IE8, though I think the problem is more IE8 specific. When starting a debug session 9 times out of 10 I will have the following problem. IE tab says connecting.. - then after a 5 second wait I will get an error in VS saying element not found. Even when I click ok to dismiss the error, the IE window still shows connecting... I will then have to kill the IE process to be able to close IE to try again. Sometimes however I am lucky and it starts. But the whole thing is so random I have no clue where to start. One thing I have noticed is that I always have 2 IE processes started even though there is only one window open. One has a small footprint of 100k, I presume it is some kind of helper. I am using a static port with the built in WebDev server. If anyone has had similar problems please let me know how you resolved it. Its driving me nuts! thanks

  Read the article

• Split a string by comma, quote and full-stop.. with a few exceptions

  - by dunc

  I've got a lot of text, similar to the following paragraph, which I'd like to split into words without punctuation (', ", ,, ., newline etc).. with a few exceptions. Initially considered endemic to the Chalakudy River system in Kerala state, southern India, but now recognised to have a wider distribution in surrounding drainages including the Periyar, Manimala, and Pamba river though the Manimala data may be questionable given it seems to be the type locality of P. denisonii. In the Achankovil River basin it occurs sympatrically, and sometimes syntopically, with P. denisonii. Wild stocks may have dwindled by as much as 50% in the last 15 years or so with collection for the aquarium trade largely held responsible although habitats are also being degraded by pollution from agricultural and domestic sources, plus destructive fishing methods involving explosives or organic toxins. The text refers to P. denisonii which is a species of fish. It's an abbreviation of Genus species. I would like this reference to be one word. So, for instance, this is the kind of array I'd like to see: Array ( ... [44] given [45] it [46] seems [47] to [48] be [49] the [50] type [51] locality [52] of [53] P. denisonii [54] In [55] the ... ) The only things that distinguish these species references such as P. denisonii from a new sentence like end. New are: The P (for Puntius, as in the P. in the aforementioned example) is only ever one letter, always a capital the d (as in . denisonii) is always either a lower case letter or an apostrophe (') What regexp can I use with preg_split to give me such an array? I've tried a simple explode( " ", $array ) but it doesn't do the job at all. Thanks in advance,

  Read the article

• JQuery event that triggers after CSS is loaded?

  - by Dave

  (I posted this on the jquery forums but it's still awaiting moderation, so I thought I'd try my luck here since stackoverflow is so awesome. If I get an answer I'll post it here.) Maybe someone can help me with this, I have a couple of links on my page (inside a <div id="theme-selector">) which allow you to change the CSS stylesheets: $('#theme-selector a').click(function(){ var path = $(this).attr('href'); $('head link').remove(); $('head').append('<link type="text/css" href="'+path+'" rel="stylesheet" />'); return false; }); Now, after I've changed the style on the page, I want to get the new background color, using the following code (which I put after the $('head').append call): var bgcolor = $('body').css('background-color'); alert(bgcolor); The problem is, I think, that it takes some time for the browser to download the new stylesheet and I sometimes get the old background color in my alert message. Is there some event I can bind that will only alert me after all the stylesheets are loaded on the page? At the moment, all I can think of is using a setTimeout(function(){}, 5000); which isn't great, because what if it takes longer/shorter to load all the CSS on the page. Let me know if I need to clarify anything and I can provide more code. Thanks in advance.

  Read the article

• Should the program logic reside inside the gui object class or be external to the class?

  - by hd112

  I have a question about how to structure code in relation to GUI objects. Suppose I have a dialog that has a list control that has a bunch of names obtained from a database. The user can edit the names. Does the logic reside inside that dialog class or should it be from the outside. To illustrate what I mean, here’s some pseudo code showing the structure of the code when the logic is handled outside the dialog class: NamesDialog : wxDialog { Private: ..stuff.. Public: ... SetNames(wxStringArray names); wxStringArray GetNames(); ..stuff.. } So the user of the class would do something like: wxStringArray names = DatabaseManager::Get()->GetNames(); names.Sort(); NamesDialogObject.SetNames(names); NamesDialogObject.ShowModal(); wxStringArray modified_names = NamesDialogObject.GetNames(); AddToDatabase(modified_names); //or something like this. On the other hand, the database logic can reside inside the NamesDialog class itself. In the show method I can query the database for the names and as the user interacts with the controls (list control in this case), the database can be updated from the event handlers. As a result the NamesDialog class only has the Show() method as there is no need to use SetNames or GetNames() etc. Which method is generally preferred? I don’t have much work experience so I’m not sure which is the proper way to handle it. Sometimes it's easier to handle everything in the class but getting access to objects it interacts with can be challenging. Generally can do it by having the relevant objects be singletons like the database manager in the above example.

  Read the article

• (closed) nodejs responding an image (via pipe and response.end()) leads into strange behaviour

  - by johannesboyne

  What the Problem was: I had 2 different writeStreams! If WriteStram#1 is closed, the second should be closed too and then it all should be piped... BUT node is asynchronious so while one has been closed, the other one hasn't. Even the stream.end() was called... well you always should wait for the close event! thx guys for your help! I am at my wit's end. I used this code to pipe an image to my clients: req.pipe(fs.createReadStream(__dirname+'/imgen/cached_images/' + link).pipe(res)) it does work but sometimes the image is not transferred completely. But no error is thrown neither on client side (browser) nor on server side (node.js). My second try was var img = fs.readFileSync(__dirname+'/imgen/cached_images/' + link); res.writeHead(200, { 'Content-Type' : 'image/png' }); res.end(img, 'binary'); but it leads to the same strange behaviour... Does anyone got a clue for me? Regards! (abstracted code...) var http = require('http'); http.createServer(function (req, res) { Imgen.generateNew( 'virtualtwins/www_leonardocampus_de/overview/28', 'www.leonardocampus.de', 'overview', '28', null, [], [], function (link) { fs.stat(__dirname+'/imgen/cached_images/' + link, function(err, file_info) { if (err) { console.log('err', err); } console.log('file info', file_info.size); res.writeHead(200, 'image/png'); fs.createReadStream(__dirname+'/imgen/cached_images/' + link).pipe(res); }); } ); }).listen(13337, '127.0.0.1'); Imgen.generateNew just creates a new file, saves it to the disk and gives back the path (link)

  Read the article

• Socket Performance C++ Or C#

  - by modernzombie

  I have to write an application that is essentially a proxy server to handle all HTTP and HTTPS requests from our server (web browsing, etc). I know very little C++ and am very comfortable writing the application features in C#. I have experimented with the proxy from Mentalis (socket proxy) which seems to work fine for small webpages but if I go to large sites like tigerdirect.ca and browse through a couple of layers it is very slow and sometimes requests don't complete and I see broken images and javascript errors. This happens with all of our vendor sites and other content heavy sites. Mentalis uses HTTP 1.0 which I know is not as efficient but should a proxy be that slow? What is an acceptable amount of performance loss from using a proxy? Would HTTP 1.1 make a noticeable difference? Would a C++ proxy be much faster than one in C#? Is the Mentalis code just not efficient? Would I be able to use a premade C++ proxy and import the DLL to C# and still get good performance or would this project call for all C++? Sorry if these are obvious questions but I have not done network programming before.

  Read the article

• Problem displaying the Message box in MFC

  - by kiddo

  I have a simple MFC program which displays the progressbar..I used the below code to display the progress bar.. HWND dialogHandle = CreateWindowEx(0,WC_DIALOG,L"Proccessing...",WS_OVERLAPPEDWINDOW|WS_VISIBLE, 600,300,280,120,NULL,NULL,NULL,NULL); HWND progressBarHandle = CreateWindowEx(NULL,PROGRESS_CLASS,NULL,WS_CHILD|WS_VISIBLE|PBS_MARQUEE,40,20,200,20, dialogHandle,(HMENU)IDD_PROGRESS,NULL,NULL); while(FALSE == testResult) { MSG msg; SendMessage(progressBarHandle, PBM_SETRANGE, 0, MAKELPARAM( 0, 100 ) ); SendMessage(progressBarHandle,PBM_SETPOS,0,0); ShowWindow(progressBarHandle,SW_SHOW); Sleep(50); if(TRUE == myCondition)//myCondition is a bool variable which is decalred globally { DestroyWindow(dialogHandle); AfxMessageBox(L"Test Success"); } } when I execute the above code..the message box displays only after a mouseover event.like if I move the mouse the message box will display if not it will not display until i move the mouse. And also while the progressbar is running if I try to move the progress bar window..it displays a windows background at the place of displacement and also in the new region or sometimes its getting stuck.Please help me with this!

  Read the article

• Application HANGS UP if 3G connection is used (WiFi is OK)

  - by SkyEagle888

  Hi, My application uses the following codes to connect to web site and download a web page for processing. HttpURLConnection c = (HttpURLConnection) url.openConnection(); c.setRequestMethod("GET"); c.setRequestProperty("User-Agent", "Mozilla/3.0 (compatible; MSIE 4.0; Windows NT)"); c.setRequestProperty("Accept-Language", "zh-hk"); c.setRequestProperty("Connection", "Keep-Alive"); c.setConnectTimeout(6000); // mTimeout is in seconds intTries = 3; do { c.connect(); if (c.getResponseCode() == 200) { break; } else { intTries --; } } while (intTries > 0); The codes are run in main thread only. If WiFi is being used, it is nice and clean. If 3G connection is being used, sometimes, it cannot get connected OR application HANGS. Is there any significant difference between handling WiFi and 3G ? Thanks.

  Read the article

< Previous Page | 280 281 282 283 284 285 286 287 288 289 290 291  | Next Page >